Load isofiles
isofiles <-
file.path(
"Results",
c(
"180226_alkanes_C_GB",
"180228_alkanes_C_GB"
)
) %>%
iso_read_continuous_flow()
## # A tibble: 12 x 4
## file_id type func details
## <chr> <chr> <chr> <chr>
## 1 511__Linearity CO2__.dxf error extract_dxf_r… cannot identify measured…
## 2 511__Linearity CO2__.dxf error extract_isoda… `peak`, `start`, `rt`, `…
## 3 512__Linearity CO2__.dxf error extract_dxf_r… cannot identify measured…
## 4 512__Linearity CO2__.dxf error extract_isoda… `peak`, `start`, `rt`, `…
## 5 513__Linearity CO2__.dxf error extract_dxf_r… cannot identify measured…
## 6 513__Linearity CO2__.dxf error extract_isoda… `peak`, `start`, `rt`, `…
## 7 535__Linearity CO2__.dxf error extract_dxf_r… cannot identify measured…
## 8 535__Linearity CO2__.dxf error extract_isoda… `peak`, `start`, `rt`, `…
## 9 536__Linearity CO2__.dxf error extract_dxf_r… cannot identify measured…
## 10 536__Linearity CO2__.dxf error extract_isoda… `peak`, `start`, `rt`, `…
## 11 537__Linearity CO2__.dxf error extract_dxf_r… cannot identify measured…
## 12 537__Linearity CO2__.dxf error extract_isoda… `peak`, `start`, `rt`, `…
Data
File Info
# all file info
isofiles %>% iso_get_file_info()
## Info: aggregating file info from 45 data file(s)
Vendor data table
isofiles %>% iso_get_vendor_data_table()
## Info: aggregating vendor data table without units from 45 data file(s)
Chromatograms
isofiles %>%
# fetch a few of the files of interst
iso_filter_files(parse_number(Analysis) %in% c(543, 550)) %>%
# plot just mass 28
iso_plot_continuous_flow_data(data = c(46)) %>%
# make interactive
ggplotly(dynamicTicks = TRUE)
## Warning: You need the dev version of ggplot2 to use `dynamicTicks`
Analysis
Select relevant data
data_table <-
isofiles %>%
# filter the files you want to use
# --> exclude CO2 zeros
iso_filter_files(!str_detect(`Identifier 1`, "CO2 zero")) %>%
# --> select only good analyses
iso_filter_files(parse_number(Analysis) > 515) %>%
# get the vendor data table and file info
iso_get_vendor_data_table(
select = c(
# peak info
Nr., Start, Rt, End,
# amplitudes and intensities
`Ampl 44`, `Ampl 46`, `Intensity All`, `Intensity 44`,
# ratios and deltas
`R 46CO2/44CO2`, `d 13C/12C`
),
include_file_info = c(
file_datetime, Analysis, `Identifier 1`, `Identifier 2`, `GC Method`, `Seed Oxidation`
)
) %>%
# rename some columns to be easier to work with
rename(
Ampl44=`Ampl 44`, Ampl46=`Ampl 46`, # amplitudes
#Area28=`rIntensity 28`, Area29=`rIntensity 29`, # areas
Intensity=`Intensity All`, #peak intensities
R46C = `R 46CO2/44CO2`, d13C = `d 13C/12C`, # ratio and delta values
rt = `Rt`,
rt_start = `Start`,
rt_end = `End`
#File = `file_id`
)
## Info: applying file filter, keeping 37 of 45 files
## Info: applying file filter, keeping 32 of 37 files
## Info: aggregating vendor data table without units from 32 data file(s), including file info 'c(file_datetime, Analysis, `Identifier 1`, `Identifier 2`, `GC Method`,
## `Seed Oxidation`)'
data_table
Map peaks
### CHANGE MAPPING FILE NAME ###
metadata_samples <- read_excel(file.path("metadata", "20180226_GB_metadata.xlsx"), sheet = "files")
metadata_peak_maps <- read_excel(file.path("metadata","20180226_GB_metadata.xlsx"), sheet = "maps")
metadata_samples
data_table_with_peaks <- data_table %>%
iso_add_metadata(metadata_samples, match_by = c(`Identifier 1`, Analysis)) %>%
iso_map_peaks(metadata_peak_maps)
## Info: metadata added to 651 data rows, 0 left without metadata:
## - 1 metadata entries were mapped to 651 data entires via column 'Identifier 1'
## Info: 618 peaks in 29 files were successfully assigned, 33 could not be assigned, and 136 were missing
# missing and unidentified peaks
data_table_with_peaks %>% filter(!is_identified | is_missing)
# confirmed peaks
data_table_with_peaks <- data_table_with_peaks %>%
filter(is_identified, !is_missing, !is_ambiguous)
data_table_with_peaks
Check stability of reference peaks
The second one is the one defined to be 0 permil (so will always be), the rest is relative to that peak.
p <- data_table_with_peaks %>%
filter(is_ref_peak == "yes") %>%
ggplot() +
aes(Nr., d13C, color = file_id) +
geom_line() +
theme_bw()
ggplotly(p)
Add standard values
standards <- read_excel(file.path("metadata", "gc_irms_indiana_A6.xlsx"))
data_w_stds <- data_table_with_peaks %>%
filter(type == "standard", is_ref_peak == "no") %>%
left_join(standards, by = "compound") %>%
mutate(is_std = !is.na(true.d13C) | !is.na(true.d2H))
Process data
Focus on the analytes and calculate a few summary parameters we want to use later.
data_w_analyte_peaks <-
data_table_with_peaks %>%
# this is important so that the reference peaks are not caught up in the next set of calculations
filter(is_ref_peak == "no") %>%
# for each analysis calculate averages across analysis
group_by(Analysis) %>%
mutate(
ampl_sample_mean.mV = mean(`Ampl44`), ampl_sample_sd.mV = sd(`Ampl44`),
area_sample_mean.Vs = mean(`Intensity 44`), area_sample_sd.Vs = sd(`Intensity 44`)
)
Standards
standards <- read_excel(file.path("metadata", "gc_irms_indiana_A6.xlsx"))
kable(standards)
| C16 |
-9.1 |
-26.15 |
| C17 |
-117.8 |
-31.88 |
| C18 |
-52.0 |
-32.70 |
| C19 |
-56.3 |
-31.99 |
| C20 |
-89.7 |
-33.97 |
| C21 |
-177.8 |
-28.83 |
| C22 |
-81.3 |
-33.77 |
| C23 |
-67.2 |
-33.37 |
| C24 |
-29.7 |
-32.13 |
| C25 |
-263.0 |
-28.46 |
| C26 |
-45.9 |
-32.94 |
| C27 |
-172.8 |
-30.49 |
| C28 |
-36.8 |
-33.20 |
| C29 |
-177.8 |
-29.10 |
| C30 |
-213.6 |
-29.84 |
data_w_stds <-
data_w_analyte_peaks %>%
iso_add_standards(standards)
## Info: added 13 standard entries to 299 out of 299 rows
Visualize standards
v <- data_w_stds %>%
ggplot() +
aes(x = true.d13C, y = d13C, color = file_id) +
geom_smooth(method = "lm", se = FALSE, alpha = 0.5) +
geom_point() +
theme_bw() +
theme(legend.position = "none")
ggplotly(v)
## We recommend that you use the dev version of ggplot2 with `ggplotly()`
## Install it with: `devtools::install_github('hadley/ggplot2')`
Calibration
data_w_calibs <- data_w_stds %>%
# prepare for calibration by defining the grouping column(s) and setting default parameters
iso_prepare_for_calibration(group_by = c(Analysis)) %>%
iso_set_default_process_parameters(delta_residual = resid_d13C) %>%
# run calibration
iso_calibrate_delta(model = lm(`d13C` ~ true.d13C)) %>%
# pull out some columns we want generally available
iso_unnest_calib_data(select = c(starts_with("Id"), ampl_sample_mean.mV))
## Info: preparing data for calibration by grouping based on 'Analysis' and nesting the grouped datasets into 'calibration_data'
## Info: calculating delta calibration fits based on 1 model ('lm(d13C ~ true.d13C)') for 26 data group(s) in 'calibration_data' with filter 'is_standard'; storing residuals in 'resid_d13C.'
data_w_calibs %>%
iso_unnest_delta_calib_summary(keep_other_list_data = FALSE) %>%
kable(digits =3)
| 517 |
A6 |
0.5uL |
397.946 |
0.870 |
0.854 |
0.859 |
53.467 |
0.000 |
2 |
-11.551 |
29.102 |
30.010 |
5.900 |
8 |
| 518 |
A6 |
1.0uL |
1028.965 |
0.853 |
0.839 |
0.956 |
63.715 |
0.000 |
2 |
-16.780 |
39.560 |
41.255 |
10.060 |
11 |
| 519 |
A6 |
2.0uL |
2044.818 |
0.932 |
0.926 |
0.591 |
151.570 |
0.000 |
2 |
-10.522 |
27.044 |
28.739 |
3.841 |
11 |
| 520 |
A6 |
3uL |
3075.854 |
0.978 |
0.976 |
0.323 |
485.383 |
0.000 |
2 |
-2.650 |
11.300 |
12.995 |
1.144 |
11 |
| 522 |
A6 |
0.5uL |
446.709 |
0.914 |
0.902 |
0.716 |
74.560 |
0.000 |
2 |
-8.637 |
23.275 |
23.866 |
3.592 |
7 |
| 523 |
A6 |
1.0uL |
1054.275 |
0.894 |
0.884 |
0.788 |
92.529 |
0.000 |
2 |
-14.265 |
34.530 |
36.225 |
6.832 |
11 |
| 524 |
A6 |
2.0uL |
2062.263 |
0.929 |
0.922 |
0.613 |
143.730 |
0.000 |
2 |
-11.005 |
28.010 |
29.705 |
4.138 |
11 |
| 525 |
A6 |
3uL |
3596.214 |
0.992 |
0.991 |
0.192 |
1324.835 |
0.000 |
2 |
4.092 |
-2.184 |
-0.489 |
0.406 |
11 |
| 526 |
A6 |
0.2uL |
57.897 |
0.905 |
0.857 |
0.616 |
18.999 |
0.049 |
2 |
-2.353 |
10.707 |
8.866 |
0.760 |
2 |
| 527 |
A6 |
0.5uL |
476.656 |
0.904 |
0.890 |
0.796 |
65.928 |
0.000 |
2 |
-9.589 |
25.179 |
25.770 |
4.438 |
7 |
| 528 |
A6 |
1.0uL |
1144.098 |
0.920 |
0.913 |
0.657 |
127.265 |
0.000 |
2 |
-11.905 |
29.810 |
31.505 |
4.752 |
11 |
| 529 |
A6 |
2.0uL |
2158.592 |
0.947 |
0.943 |
0.513 |
198.511 |
0.000 |
2 |
-8.686 |
23.373 |
25.068 |
2.896 |
11 |
| 530 |
A6 |
3uL |
3608.378 |
0.995 |
0.995 |
0.142 |
2409.987 |
0.000 |
2 |
8.019 |
-10.038 |
-8.343 |
0.222 |
11 |
| 540 |
A6 |
0.5uL |
423.750 |
0.886 |
0.870 |
0.849 |
54.634 |
0.000 |
2 |
-10.165 |
26.330 |
26.921 |
5.044 |
7 |
| 541 |
A6 |
0.5uL |
412.138 |
0.899 |
0.884 |
0.822 |
62.157 |
0.000 |
2 |
-9.876 |
25.752 |
26.343 |
4.730 |
7 |
| 542 |
A6 |
0.5uL |
413.941 |
0.872 |
0.856 |
0.941 |
54.685 |
0.000 |
2 |
-12.463 |
30.926 |
31.834 |
7.080 |
8 |
| 543 |
A6 |
0.5uL |
405.228 |
0.751 |
0.723 |
1.492 |
27.121 |
0.001 |
2 |
-18.905 |
43.811 |
45.005 |
20.033 |
9 |
| 544 |
A6 |
0.5uL |
441.497 |
0.867 |
0.851 |
0.922 |
52.288 |
0.000 |
2 |
-12.261 |
30.522 |
31.430 |
6.800 |
8 |
| 545 |
A6 |
0.5uL |
413.272 |
0.832 |
0.817 |
1.085 |
54.406 |
0.000 |
2 |
-18.420 |
42.840 |
44.535 |
12.948 |
11 |
| 546 |
A6 |
0.5uL |
445.801 |
0.853 |
0.834 |
0.967 |
46.318 |
0.000 |
2 |
-12.734 |
31.469 |
32.377 |
7.475 |
8 |
| 547 |
A6 |
1.0uL |
1161.397 |
0.879 |
0.868 |
0.861 |
79.857 |
0.000 |
2 |
-15.411 |
36.821 |
38.516 |
8.149 |
11 |
| 548 |
A6 |
1.0uL |
1162.422 |
0.865 |
0.852 |
0.894 |
70.351 |
0.000 |
2 |
-15.901 |
37.802 |
39.497 |
8.788 |
11 |
| 549 |
A6 |
1.0uL |
1219.784 |
0.898 |
0.888 |
0.757 |
96.471 |
0.000 |
2 |
-13.733 |
33.465 |
35.160 |
6.295 |
11 |
| 550 |
A6 |
1.0uL |
1288.446 |
0.889 |
0.879 |
0.787 |
88.190 |
0.000 |
2 |
-14.252 |
34.504 |
36.199 |
6.819 |
11 |
| 551 |
A6 |
1.0uL |
1292.487 |
0.915 |
0.907 |
0.685 |
117.820 |
0.000 |
2 |
-12.433 |
30.865 |
32.560 |
5.154 |
11 |
| 552 |
A6 |
1.0uL |
1381.540 |
0.898 |
0.888 |
0.754 |
96.335 |
0.000 |
2 |
-13.695 |
33.389 |
35.084 |
6.259 |
11 |
data_w_calibs %>%
# pull out seed oxidation in addition
iso_unnest_calib_data(select = `Seed Oxidation`) %>%
iso_unnest_delta_calib_coefs(select = c(-statistic), keep_other_list_data = FALSE) %>%
# arrange by term and Analysis to get a quick idea of what the numbers across analyses
arrange(term, Analysis) %>%
kable(digits = 2)
| 517 |
A6 |
0.5uL |
397.95 |
1 |
(Intercept) |
11.08 |
4.46 |
0.04 |
* |
| 518 |
A6 |
1.0uL |
1028.96 |
1 |
(Intercept) |
13.21 |
4.39 |
0.01 |
* |
| 519 |
A6 |
2.0uL |
2044.82 |
1 |
(Intercept) |
11.91 |
2.71 |
0.00 |
** |
| 520 |
A6 |
3uL |
3075.85 |
1 |
(Intercept) |
11.22 |
1.48 |
0.00 |
*** |
| 522 |
A6 |
0.5uL |
446.71 |
1 |
(Intercept) |
13.68 |
4.09 |
0.01 |
* |
| 523 |
A6 |
1.0uL |
1054.27 |
1 |
(Intercept) |
13.10 |
3.62 |
0.00 |
** |
| 524 |
A6 |
2.0uL |
2062.26 |
1 |
(Intercept) |
12.23 |
2.82 |
0.00 |
** |
| 525 |
A6 |
3uL |
3596.21 |
1 |
(Intercept) |
10.72 |
0.88 |
0.00 |
*** |
| 526 |
A6 |
0.2uL |
57.90 |
1 |
(Intercept) |
5.98 |
6.11 |
0.43 |
|
| 527 |
A6 |
0.5uL |
476.66 |
1 |
(Intercept) |
15.45 |
4.54 |
0.01 |
* |
| 528 |
A6 |
1.0uL |
1144.10 |
1 |
(Intercept) |
12.47 |
3.02 |
0.00 |
** |
| 529 |
A6 |
2.0uL |
2158.59 |
1 |
(Intercept) |
11.81 |
2.36 |
0.00 |
*** |
| 530 |
A6 |
3uL |
3608.38 |
1 |
(Intercept) |
10.74 |
0.65 |
0.00 |
*** |
| 540 |
A6 |
0.5uL |
423.75 |
1 |
(Intercept) |
14.20 |
4.84 |
0.02 |
* |
| 541 |
A6 |
0.5uL |
412.14 |
1 |
(Intercept) |
15.43 |
4.69 |
0.01 |
* |
| 542 |
A6 |
0.5uL |
413.94 |
1 |
(Intercept) |
14.58 |
4.89 |
0.02 |
* |
| 543 |
A6 |
0.5uL |
405.23 |
1 |
(Intercept) |
17.35 |
7.54 |
0.05 |
* |
| 544 |
A6 |
0.5uL |
441.50 |
1 |
(Intercept) |
13.18 |
4.79 |
0.02 |
* |
| 545 |
A6 |
0.5uL |
413.27 |
1 |
(Intercept) |
14.54 |
4.98 |
0.01 |
* |
| 546 |
A6 |
0.5uL |
445.80 |
1 |
(Intercept) |
12.79 |
5.02 |
0.03 |
* |
| 547 |
A6 |
1.0uL |
1161.40 |
1 |
(Intercept) |
13.68 |
3.95 |
0.01 |
** |
| 548 |
A6 |
1.0uL |
1162.42 |
1 |
(Intercept) |
12.93 |
4.10 |
0.01 |
** |
| 549 |
A6 |
1.0uL |
1219.78 |
1 |
(Intercept) |
12.63 |
3.47 |
0.00 |
** |
| 550 |
A6 |
1.0uL |
1288.45 |
1 |
(Intercept) |
12.45 |
3.61 |
0.01 |
** |
| 551 |
A6 |
1.0uL |
1292.49 |
1 |
(Intercept) |
12.67 |
3.14 |
0.00 |
** |
| 552 |
A6 |
1.0uL |
1381.54 |
1 |
(Intercept) |
12.48 |
3.46 |
0.00 |
** |
| 517 |
A6 |
0.5uL |
397.95 |
1 |
true.d13C |
1.05 |
0.14 |
0.00 |
*** |
| 518 |
A6 |
1.0uL |
1028.96 |
1 |
true.d13C |
1.11 |
0.14 |
0.00 |
*** |
| 519 |
A6 |
2.0uL |
2044.82 |
1 |
true.d13C |
1.05 |
0.09 |
0.00 |
*** |
| 520 |
A6 |
3uL |
3075.85 |
1 |
true.d13C |
1.03 |
0.05 |
0.00 |
*** |
| 522 |
A6 |
0.5uL |
446.71 |
1 |
true.d13C |
1.12 |
0.13 |
0.00 |
*** |
| 523 |
A6 |
1.0uL |
1054.27 |
1 |
true.d13C |
1.10 |
0.11 |
0.00 |
*** |
| 524 |
A6 |
2.0uL |
2062.26 |
1 |
true.d13C |
1.07 |
0.09 |
0.00 |
*** |
| 525 |
A6 |
3uL |
3596.21 |
1 |
true.d13C |
1.01 |
0.03 |
0.00 |
*** |
| 526 |
A6 |
0.2uL |
57.90 |
1 |
true.d13C |
0.87 |
0.20 |
0.05 |
* |
| 527 |
A6 |
0.5uL |
476.66 |
1 |
true.d13C |
1.18 |
0.14 |
0.00 |
*** |
| 528 |
A6 |
1.0uL |
1144.10 |
1 |
true.d13C |
1.08 |
0.10 |
0.00 |
*** |
| 529 |
A6 |
2.0uL |
2158.59 |
1 |
true.d13C |
1.05 |
0.07 |
0.00 |
*** |
| 530 |
A6 |
3uL |
3608.38 |
1 |
true.d13C |
1.01 |
0.02 |
0.00 |
*** |
| 540 |
A6 |
0.5uL |
423.75 |
1 |
true.d13C |
1.14 |
0.15 |
0.00 |
*** |
| 541 |
A6 |
0.5uL |
412.14 |
1 |
true.d13C |
1.18 |
0.15 |
0.00 |
*** |
| 542 |
A6 |
0.5uL |
413.94 |
1 |
true.d13C |
1.16 |
0.16 |
0.00 |
*** |
| 543 |
A6 |
0.5uL |
405.23 |
1 |
true.d13C |
1.26 |
0.24 |
0.00 |
*** |
| 544 |
A6 |
0.5uL |
441.50 |
1 |
true.d13C |
1.11 |
0.15 |
0.00 |
*** |
| 545 |
A6 |
0.5uL |
413.27 |
1 |
true.d13C |
1.16 |
0.16 |
0.00 |
*** |
| 546 |
A6 |
0.5uL |
445.80 |
1 |
true.d13C |
1.10 |
0.16 |
0.00 |
*** |
| 547 |
A6 |
1.0uL |
1161.40 |
1 |
true.d13C |
1.12 |
0.12 |
0.00 |
*** |
| 548 |
A6 |
1.0uL |
1162.42 |
1 |
true.d13C |
1.09 |
0.13 |
0.00 |
*** |
| 549 |
A6 |
1.0uL |
1219.78 |
1 |
true.d13C |
1.08 |
0.11 |
0.00 |
*** |
| 550 |
A6 |
1.0uL |
1288.45 |
1 |
true.d13C |
1.07 |
0.11 |
0.00 |
*** |
| 551 |
A6 |
1.0uL |
1292.49 |
1 |
true.d13C |
1.08 |
0.10 |
0.00 |
*** |
| 552 |
A6 |
1.0uL |
1381.54 |
1 |
true.d13C |
1.07 |
0.11 |
0.00 |
*** |
Parameters
data_params <- data_w_calibs %>%
# pull out remaining columns we want available (some might already be pulled out but that's okay)
#Note: Garrett changed "Preparation" to "Seed Oxidation" - we may want a different shape variable
iso_unnest_calib_data(select = c(file_datetime, ampl_sample_mean.mV, `Seed Oxidation`, is_standard))
# visualize the delta calibration fits
data_params %>%
#NEED TO FILTER OUT SAMPLES iso_filter_files(`Identifier 1` == "A5")
iso_visualize_delta_calib_fits(x = Analysis, color = `Identifier 2`, shape = `Seed Oxidation`, size = ampl_sample_mean.mV,
include_from_summary = c(adj.r.squared, deviance)) + labs(title = "parameters vs. analysis")

data_params %>%
iso_visualize_delta_calib_fits(x = ampl_sample_mean.mV, color = `Identifier 2`, shape = `Seed Oxidation`, size = ampl_sample_mean.mV,
include_from_summary = c(adj.r.squared, deviance)) + labs(title = "parameters vs. amplitude")

data_params %>%
iso_visualize_delta_calib_fits(x = file_datetime, color = `Identifier 2`, shape = `Seed Oxidation`, size = ampl_sample_mean.mV,
include_from_summary = c(adj.r.squared, deviance)) + labs(title = "parameters vs. time")

# turn the last plot into an interactive one
ggplotly(ggplot2::last_plot() + theme(legend.position = "none"))
## We recommend that you use the dev version of ggplot2 with `ggplotly()`
## Install it with: `devtools::install_github('hadley/ggplot2')`
Compounds
data_w_calibs %>%
# pull out relevant data
iso_unnest_calib_data(select = everything()) %>%
filter(is_standard) %>%
# calculate deviation from means
group_by(Analysis) %>%
mutate(
`Var: residual d13C [permil]` = resid_d13C,
`Var: area diff from mean [%]` = (`Intensity 44`/mean(`Intensity 44`) - 1) * 100,
`Var: amplitude diff from mean [%]` = (`Ampl44`/mean(`Ampl44`) - 1) * 100
) %>%
# visualize
iso_visualize_data(x = compound, y = starts_with("Var"), group = Analysis, color = `Identifier 2`)

ggplotly(ggplot2::last_plot())
## We recommend that you use the dev version of ggplot2 with `ggplotly()`
## Install it with: `devtools::install_github('hadley/ggplot2')`